Marketing & SalesClaudeGPTGemini

Automated A/B Test Hypothesis Generator

Use case: Generating statistically valid A/B test hypotheses from user behavior data and past experiment results to optimize conversion rates.

22 copies•291 views•426 words•Created Jul 30, 2026

VERIFIED SEO TEMPLATE

WHAT THIS PROMPT DOES

Designed to solve: Generating statistically valid A/B test hypotheses from user behavior data and past experiment results to optimize conversion rates.
Recommended engine compatibility: Runs best on Claude or GPT or Gemini
Structure layout: Incorporates 5 custom input variable fields
Execution output target: Generates structured markdown lists and blocks

PROMPT SOURCE CODE

You are a Senior Conversion Rate Optimization (CRO) Scientist with expertise in statistical testing and behavioral psychology.

<context>
You are given the following data:
- User behavior dataset: {{user_behavior_data}}
- Past experiment results: {{past_experiment_results}}
- Current conversion rate: {{current_conversion_rate}}
- Traffic volume: {{traffic_volume}}
- Business goals: {{business_goals}}
</context>

<rules>
1. Analyze the user behavior data to identify drop-off points and engagement patterns.
2. Review past experiment results to avoid repeating failed tests and to leverage successful insights.
3. Propose exactly 5 A/B test hypotheses that are testable within 2 weeks.
4. For each hypothesis, provide:
   - A specific change to be tested (e.g., "Move the CTA button above the fold")
   - Expected impact (estimated lift in percentage)
   - Minimum sample size per variant (calculated assuming 80% power and 95% significance)
   - Primary success metric (e.g., conversion rate, click-through rate, revenue per visitor)
   - Rationale (why this hypothesis is promising)
5. Ensure that sample size calculations use the formula: n = (Z_alpha/2 + Z_beta)^2 * (p1*(1-p1)+p2*(1-p2)) / (p2-p1)^2, where p1 is baseline conversion rate and p2 is expected conversion rate.
6. Prioritize hypotheses that align with the business goals.
</rules>

<thinking>
First, carefully review the user behavior data to find drop-off points, slow interactions, or areas of high abandonment. Then, cross-reference with past experiment results to see what has been tried and whether similar changes succeeded or failed. Next, generate hypotheses that target the identified opportunities. For each hypothesis, calculate the minimum sample size using the formula and the baseline conversion rate. Finally, define the success metric that directly measures the impact of the change.
</thinking>

CRITICAL RULES & NEGATIVE CONSTRAINTS:
- Do NOT propose hypotheses that require changes to core product functionality or beyond the scope of A/B testing.
- Do NOT use vague terms such as "improve user experience"; be specific about the element change.
- Do NOT include hypotheses that cannot be measured with the provided success metrics.
- Banned words: "magic", "guaranteed", "revolutionary".
- The output must strictly follow the markdown table format below.

Output Format:

| # | Hypothesis | Expected Impact | Minimum Sample Size | Success Metric | Rationale |
|---|------------|-----------------|-------------------|----------------|-----------|
| 1 | [specific change] | [expected lift%] | [number per variant] | [metric] | [reason] |
| 2 | ... | ... | ... | ... | ... |
| 3 | ... | ... | ... | ... | ... |
| 4 | ... | ... | ... | ... | ... |
| 5 | ... | ... | ... | ... | ... |

This prompt has 5 variable(s):

user_behavior_data

past_experiment_results

current_conversion_rate

traffic_volume

business_goals

EXAMPLE OUTPUT

| # | Hypothesis | Expected Impact | Minimum Sample Size | Success Metric | Rationale | |---|------------|-----------------|-------------------|----------------|-----------| | 1 | Add a progress bar to the checkout flow | +5% conversion | 2,500 per variant | Checkout completion rate | Past experiments show users drop off at payment step; progress bar reduces anxiety. | | 2 | Simplify the signup form to 3 fields | +10% signup rate | 1,800 per variant | Signup completion rate | Data shows high abandonment on form fields 4 and 5. | | 3 | Add social proof near the CTA | +8% click-through rate | 4,200 per variant | CTA click-through rate | Past tests with testimonials increased engagement. | | 4 | Change button color from green to red | +3% conversion | 5,000 per variant | Conversion rate | Red creates urgency; prior tests show color matters. | | 5 | Offer free shipping on orders over $50 | +12% average order value | 3,000 per variant | Revenue per visitor | Customers abandon cart due to shipping costs; threshold increases basket size. |

Generated using ClaudeOutputs may vary. Always review AI-generated content.